Directed evolution holds promise for creating designer enzymes for use as reagents in biomedical research, medical diagnostics, and perhaps even as therapeutics. Directed evolution has been used successfully to select RNA molecules with de novo binding and catalytic properties and most recently proteins with de novo binding properties from large libraries (>108) of molecules. Researchers have not been able to bring the power of directed evolution to bear on enzyme catalysis, however, because there are no suitable high- throughput selections for function for most enzyme-catalyzed reactions. While directed evolution has now been used routinely to increase the activity of known enzymes from small libraries (103-104) of protein variants by automation of traditional enzyme assays, the routine directed evolution of enzymes with prescribed changes in substrate specificity is yet to be achieved. The de novo directed evolution of enzyme activity has never been achieved. To address this bottleneck in the directed evolution field, my laboratory created "chemical complementation" in the previous granting period[unreadable]a high-throughput assay for enzyme catalysis of bond formation and cleavage reactions. Our goal in this next granting period is to use chemical complementation to cross this next milestone in the protein directed evolution field[unreadable]the routine directed evolution of enzymes with prescribed changes in substrate specificity. Specifically, we propose to achieve this goal through the directed evolution of TIM barrels, a "privileged" scaffold for enzyme catalysis, with altered substrate specificities for catalysis of two synthetically important transformations, carbohydrate synthesis (Aim 1) and the aldol reaction (Aim 2). There are two steps to directed evolution[unreadable]first DNA mutagenesis to create large libraries of protein variants; and second, selection of the fittest variants. Having engineered yeast to expand the range of chemistry that can give the cell a selective advantage (chemical complementation), in Aim 3 we further engineer yeast to also carry out the DNA mutagenesis in vivo. Our long-term goal is the routine de novo directed evolution of TIM barrel enzymes for myriad chemical transformations.